AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

When Active Learning Meets Implicit Semantic Data Augmentation

Zhuangzhuang Chen, Jin Zhang, Pan Wang, Jie Chen, Jianqiang Li

ECCV 2022 · Label-efficient and robust learning › Semi-supervised learning · approach: Pseudo-labeling for specialized tasks · 17 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add EPMCM selection plus ISDA (DAST-AL) versus random with ISDA (ablation)Ran+ · CIFAR-10 (ResNet-18), 10% labeledaccuracy: 80.76 -> 82.98Improved
add EPMCM selection plus ISDA versus random with ISDA (ablation)Ran+ · CIFAR-10 (ResNet-18), 35% labeledaccuracy: 91.15 -> 93.32Improved
add EPMCM selection plus ISDA versus max-probability with ISDA (ablation)Maxp+ · CIFAR-10 (ResNet-18), 20% labeledaccuracy: 81.72 -> 90.42Improved
add ISDA to random selection (ablation)Ran · CIFAR-10 (ResNet-18), 10% labeledaccuracy: 80.06 -> 80.76Improved
add ISDA to random selection (ablation)Ran · CIFAR-10 (ResNet-18), 40% labeledaccuracy: 91.21 -> 91.15Hurt
add ISDA to random selection at 5% labeled (ablation)Ran · CIFAR-10 (ResNet-18), 5% labeledaccuracy: 67.13 -> 69.27Improved
add ISDA to max-probability selection (ablation)Maxp · CIFAR-10 (ResNet-18), 10% labeledaccuracy: 76.37 -> 76.35No clear effect
use max-probability selection instead of random (ablation)Ran · CIFAR-10 (ResNet-18), 35% labeledaccuracy: 91.21 -> 84.60Hurt
increase number of explicit augmented samples from 100 to 100000 in the upper-bound gap (ablation)EPMCM upper bound · CIFAR-10 ResNet18 gapgap: 0.85 ± 0.49 -> 0.11 ± 0.07Improved
use DAST-AL instead of ADS for first-iteration timeADS · CIFAR-10 timingtotal time (min): 44.91 -> 2.26Improved
use DAST-AL instead of SRAAL for AL iteration timeSRAAL · CIFAR-10 timingtotal time (min): 298.9 -> 2.26Improved
use DAST-AL instead of ADS at 2% labeled dataADS · CIFAR-10 VGG-16 (2% labeled)accuracyHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy